Page 1 of 1

treble in paradise

Posted: 14 Sep 2014, 11:16
by Fonotec
The cutting gear made by Ortofon includes treble limiters in both electron-valve and transistor topologies. These treble-limiters are not dynamics processors having three stages of limiting, however. (; They are single-stage limiters that can be made to react to only the components of the complex signals that are at thrice the frequencies of the midrange signals. The traditional midrange frequency used for tape recorder calibration is 700 Hz (and, later, they started using 1,000 Hz). As long as we are talking about frequencies over 2 kHz, we are safely in treble territory.

With the Ortofon variety of acceleration limiter, you will be transported to treble Paradise...

Image

I finally sorted out what was wrong with the left-right channel balance on the STL631 treble limiter that I acquired a couple of years ago from a colleague, who did warn me that it "needs work." The thing turned on, and the needles moved to the right, but the output channels were not matched in level, and yet there was no way to trim-adjust the outputs. Although I renewed the tubes, I held little hope of solving the matching issue, since the mismatched output levels were apparent even when the filter was bypassed and the signals were just passing through the unit - whether off, or on.

Examining the signal path as drawn in the schematic revealed that, when the filter is bypassed, both channels only send the input through a 1 kOhm resistor before hand-off to the output. When the filter is switched in, additional components are introduced, but those seemed to have no worsening effect on the channel mismatch. Left was always down by about 8 dB, compared to right. However, the schematic solved the issue. Something had to be wrong with (at least) one of the two 1 kOhm resistors, or another source of resistance was somehow in the left channel that shouldn't be there.

After sliding off the mu-metal cover, which has sliding snake connectors for the chassis' program and side-chain inputs, program outputs, and AC power inputs, I noticed a new-looking 1 kOhm resistor soldered between two of the riveted through-hole stand-offs. (Alas, the board has no pin-designating legend, but the Fluke has a convenient, audible continuity test mode.) Unfortunately, that new resistor, which was obviously soldered "in-the-field," measured pretty close to the schematic at 1.07 kOhm. But I wasn't sure which of the two channels it was for... When I turned the unit up-side-down, I saw the mirror-image of most of the layout of components from the first side. However, there was not a 1 kOhm resistor in the corresponding position. There was an old, carbon film 3.3 kOhm resistor. It measured 3.5 kOhms and must have been dropping the volts on the left channel, compared to the right. which had the 1.07 kOhm resistor (only) in line. Perhaps, the person who repaired this many decades ago was about to install the proper value resistor when the door bell or telephone rang. By the time she got back to the soldering, she accidentally grabbed a nearby resistor, thinking it was the same one she had placed next to the work. Or some M/S-only mod.

Image

After I replaced both of those resistors with two cherry-picked, new, 1.02 kOhm carbon film resitors that I had in inventory, the channel-matching became perfect.

Image

Image

[The photo, above, of the bottom breadboard of the limiter has been horizontally-reversed to make the symmetry of the two boards more obvious. Also, the precise measurement of these resistors was 1.020 kOhm and 1.022 kOhm.]

The only remaining issue seemed to be that the treble response might be slightly drooped. A 10 kHz sinewave soundfile was playing out of the limiter at about - 4 dB compared to the 500 Hz sinewave soundfile... When I played music through the unit, it sounded nice, and much improved from before, but obviously had some high end roll-off, although it was matching on both channels, even though the side-chain inputs were not receiving a signal. Alas, the overall response from bass through treble is designed to be within +/- 3 dB from true, so that might have been in spec, but I won't be surprised to find several parts being in need of replacement. The next time I get a chance, I'll try a sweep of tones and see what's going on.



Image

Image
(right-click "View Image" or "Open image in new tab" to see entire picture)

[STL631 schematic scans from etec.dk]

Re: treble in paradise

Posted: 11 Sep 2019, 22:34
by Fonotec
A thread about treble limiters from 2006 on ProSoundWeb includes a post by Tony Batcherlor who says not to use the STL-631 for protection of one of the Ortofon stereo cutting heads when being driven by modern, solid state amps (such as the GO 741).

See screengrab:
Screen Shot 2019-09-11 at 10.28.38 PM.png
Screen Shot 2019-09-11 at 10.28.38 PM.png (105.62 KiB) Viewed 5429 times
I just sold this unit to someone on ebay who paid the ceremonial $2121.21 Buy-It-Now price.

Hope he doesn't turn it into an 'audiophile' preamp.